Resumo

MARULANDA-DURANGO, Jesser J.; SEPULVEDA-LONDONO, Christian D.  e  ALVAREZ-LOPEZ, Mauricio A.. Parameter Estimation for an Electric Arc Furnace Model Using Maximum Likelihood. TecnoL. [online]. 2012, n.29, pp.69-89. ISSN 0123-7799.

In this paper, we present a methodology for estimating the parameters of a model for an electrical arc furnace, by using maximum likelihood estimation. Maximum likelihood estimation is one of the most employed methods for parameter estimation in practical settings. The model for the electrical arc furnace that we consider, takes into account the non-periodic and non-linear variations in the voltage-current characteristic. We use NETLAB, an open source MATLAB® toolbox, for solving a set of non-linear algebraic equations that relate all the parameters to be estimated. Results obtained through simulation of the model in PSCADTM, are contrasted against real measurements taken during the furnance's most critical operating point. We show how the model for the electrical arc furnace, with appropriate parameter tuning, captures with great detail the real voltage and current waveforms generated by the system. Results obtained show a maximum error of 5% for the current's root mean square error.

Palavras-chave : Arc furnace; harmonics; dynamic models; maximum likelihood estimation.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )